CN109917902A - A kind of server power-economizing method, cluster management controller and energy-saving server - Google Patents
A kind of server power-economizing method, cluster management controller and energy-saving server Download PDFInfo
- Publication number
- CN109917902A CN109917902A CN201910213121.7A CN201910213121A CN109917902A CN 109917902 A CN109917902 A CN 109917902A CN 201910213121 A CN201910213121 A CN 201910213121A CN 109917902 A CN109917902 A CN 109917902A
- Authority
- CN
- China
- Prior art keywords
- server
- nvram
- cluster
- management controller
- command
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Power Sources (AREA)
Abstract
The application discloses a kind of server power-economizing method, cluster management controller and energy-saving server.This method sends the first control command from cluster management controller to first server, and first server stores the initiation parameter of first server memory into NVRAM according to the first control command.Cluster management controller, which determines, sends shutdown command to first server when first server is in idling conditions, controls first server from idling conditions and switches to off-mode, switches the power consumption for saving first server by state.NVRAM is still able to maintain data after a loss of power, when first server needs restart under off-mode, the initiation parameter that memory can be directly read from NVRAM, effectively shortens initialization time of first server memory, promotes the starting speed of first server.The power-economizing method is not only energy saving, can also make server quick start after energy conservation shutdown, promote the working efficiency of server.
Description
Technical field
This application involves field of communication technology, more particularly to a kind of server power-economizing method, cluster management controller and
Energy-saving server.
Background technique
With the increase in demand of marketing data, the demand of data center is more and more.In the data center, the number of server
How much amount determines the traffic handing capacity of this data center.Currently, the quantity of server is logical in medium-and-large-sized data center
Often reach thousands of.The server apparatus of flood tide in this way operates simultaneously, calculates according to every server operation power is 500W,
10000 will 5,000 kW.As it can be seen that the power consumption of a data center is very large.
Server has several big building blocks: central processing unit (Central Processing Unit, CPU), memory, magnetic
Disk, network interface card and main board elements etc..Wherein, CPU power consumption accounts for the one third of entire server energy consumption, and the computing capability of CPU is got over
By force, power consumption is bigger.Even if CPU can also reach the half of rated disspation under server idling conditions.Memory is clothes
The component that most consumes energy in addition to CPU in business device, at present the forth generation double data rate synchronous dynamic random access memory DDR4 memory of mainstream
Several watts magnitude is had reached, clocked memory is higher, and power consumption is higher.Since a server generally configures tens memories, cause whole
The energy consumption accounting of a memory is also bigger.The peripheral equipments such as disk and network interface card can also consume a part of power consumption.
But the mission requirements of server have regularity, whole server few in portion of time server task
In idle running (IDLE) state.Even if under IDEL state, the power consumption of server complete machine is also up to up to a hundred watts.
Summary of the invention
Based on the above issues, this application provides a kind of server power-economizing method, cluster management controller and energy conservation services
Device saves the power consumption of data center to save server energy consumption.
The embodiment of the present application discloses following technical solution:
The application is in a first aspect, provide a kind of server power-economizing method, applied to cluster management controller, the method packet
It includes:
The first control command is sent to the first server;First control command is used to indicate first server will
The initiation parameter of the memory of the first server is stored into nonvolatile random access memory NVRAM;Described first
Server and the NVRAM are connected with each other;
When determining that the first server is in idling conditions, Xiang Suoshu first server sends shutdown command.
Optionally, before the determination server is in idling conditions, the method also includes:
Obtain the corresponding data of each server in cluster;Each server includes the first server;
The corresponding data of each server in the cluster are analyzed, the task of the first server are transferred to described each
Other servers in a server.
Optionally, after the transmission shutdown command to the first server, the method also includes:
Obtain the mission requirements in cluster;
According to the mission requirements, Xiang Suoshu first server sends start command, so as to first server response
In the start command, starting is completed after reading the initiation parameter in the NVRAM.
The application second aspect provides a kind of cluster management controller, comprising:
First sending module, for sending the first control command to the first server;First control command is used
The initiation parameter of the memory of the first server is stored to based non-volatile random access in instruction first server and is stored
In device NVRAM;The first server and the NVRAM are connected with each other;
Second sending module, when for determining that the first server is in idling conditions, Xiang Suoshu first server hair
Send shutdown command.
Optionally, cluster management controller further include:
First obtains module, for obtaining the corresponding data of each server in cluster;Each server includes institute
State first server;
Data analysis module, for analyzing the corresponding data of each server in the cluster;
Task shift module, other clothes for being transferred to the task of the first server in each server
Business device.
Optionally, cluster management controller further include:
Second obtains module, for obtaining the mission requirements in cluster;
Third sending module, for according to the mission requirements, Xiang Suoshu first server to send start command, with toilet
First server is stated in response to the start command, starting is completed after reading the initiation parameter in the NVRAM.
The application third aspect provides a kind of server power-economizing method, is applied to server, which comprises
Receive the first control command;
The initiation parameter of the memory of the server is stored to non-volatile random according to first control command
It accesses in memory NVRAM;The server and the NVRAM are connected with each other;
Shutdown command is received when the server is in idling conditions, according to the shutdown command by the server
All building blocks are closed.
Optionally, before the server is in idling conditions, the method also includes:
The corresponding data of the server are sent to cluster management controller, so that the cluster management controller is according to institute
It states the corresponding data of server to be analyzed, and the task of the server is transferred to other servers in cluster.
Optionally, it is described all building blocks of the server are closed according to the shutdown command after, it is described
Method further include:
Receive start command;The start command is used to indicate the server starting;
In response to the start command, the initiation parameter, the server starting are read from the NVRAM.
The application fourth aspect provides a kind of energy-saving server, comprising:
First receiving unit, for receiving the first control command;
Data storage cell, for being deposited the initiation parameter of the memory of the server according to first control command
Storage is into nonvolatile random access memory NVRAM;The server and the NVRAM are connected with each other;
Second receiving unit, for receiving shutdown command when the server is in idling conditions;
Shut down unit, for being closed all building blocks of the server according to the shutdown command.
Optionally, server further include:
First transmission unit, for sending the corresponding data of the server to cluster management controller, so as to the collection
Group's Management Controller is analyzed according to the corresponding data of the server, and the task of the server is transferred in cluster
Other servers.
Optionally, server further include:
Third receiving unit, for receiving start command;The start command is used to indicate the server starting;
Start unit, it is described for reading the initiation parameter from the NVRAM in response to the start command
Server starting.
Compared to the prior art, the application has the advantages that
Server power-economizing method provided by the embodiments of the present application sends first from cluster management controller to first server
Control command, first server store the initiation parameter of first server memory to non-volatile according to the first control command
In random access storage device NVRAM.When cluster management controller determines that first server is in idling conditions, to first server
Shutdown command is sent, off-mode is switched to from idling conditions to control first server, is switched by state, saves first
The power consumption of server.
Since NVRAM is still able to maintain data after a loss of power, when first server needs the weight under off-mode
When new starting, first server can directly read the initiation parameter of memory from NVRAM, effectively shorten first service in this way
The initialization time of device memory promotes the starting speed of first server.It can be seen that the power-economizing method is not only energy saving, may be used also
So that server quick start after energy conservation shutdown, promotes the working efficiency of server.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application without any creative labor, may be used also for those of ordinary skill in the art
To obtain other drawings based on these drawings.
Fig. 1 is a kind of flow chart of server power-economizing method provided by the embodiments of the present application;
Fig. 2 is a kind of application scenarios schematic diagram of server power-economizing method provided by the embodiments of the present application;
Fig. 3 is the flow chart of another server power-economizing method provided by the embodiments of the present application;
Fig. 4 is a kind of structural schematic diagram of cluster management controller provided by the embodiments of the present application;
Fig. 5 is the flow chart of another server power-economizing method provided by the embodiments of the present application;
Fig. 6 is the flow chart of another server power-economizing method provided by the embodiments of the present application;
Fig. 7 is a kind of structural schematic diagram of energy-saving server provided by the embodiments of the present application.
Specific embodiment
It describes as discussed above, the number of servers of data center configuration is huge at present, and the power consumption burden of data center is extremely heavy.
Even under idling conditions, every server still keeps the power consumption of hectowatt.
By research, inventors have found that can control the server shutdown of idle running, to further save the function of server
Consumption.But the restarting process of server is very slow, once server shuts down, that is, is difficult to quick response service.
Inventors discovered through research that the initialization time of server memory occupies the 80% of server start-up course time
Left and right.Based on this, inventor proposes a kind of server power-economizing method, cluster management controller and energy-saving server, by that will take
The initiation parameter of business device memory is stored to the mode of NVRAM, is quickly restarted with realizing that server energy conservation is postboost.
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only this
Invention a part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art exist
Every other embodiment obtained under the premise of creative work is not made, shall fall within the protection scope of the present invention.
First embodiment
Referring to Fig. 1, which is a kind of flow chart of server power-economizing method provided by the embodiments of the present application.It needs to illustrate
It is that server power-economizing method provided in this embodiment is applied to cluster management controller.
As shown in Figure 1, server power-economizing method includes:
Step 101: Xiang Suoshu first server sends the first control command, and first control command is used to indicate first
Server stores the initiation parameter of the memory of the first server into nonvolatile random access memory NVRAM.
For ease of understanding, reference can be made to Fig. 2, the figure are the application scenarios signal of server power-economizing method provided in this embodiment
Figure.It include multiple nodes, i.e. first server 201, second server 202, third server 203 etc. in cluster in the scene.
Three servers are exemplarily only provided in figure, in practical applications, in cluster in the application scenarios of the present embodiment method
Number of servers is not limited.Each node is controlled and is managed by cluster management controller 204 respectively in cluster.It can be with
Understand, each server can communicate respectively with the realization of cluster management controller 204 and data are transmitted.
It is below example with first server 201, the specific implementation of server power-economizing method is described in detail.
To start convenient for can rapidly be realized according to the demand of data center after 201 energy conservation shutdown of first server, this
In step, the first control command is sent from cluster management controller 204 to first server 201.When first server 201 receives
To after first control command, the initiation parameter of the memory of first server 201 can be stored to according to the order
In the NVRAM that one server 201 is connected.Because NVRAM is a kind of memory that can keep data in the power-off state, because
The internal memory initialization parameter of this deposit NVRAM is taken at any time for first server 201.
It is understood that in this step, the initiation parameter deposit NVRAM of memory is by control first server 201
An energy-efficient important prerequisite is realized in the present embodiment.This is because, if not executing step 101 and only executing step in isolation
102, even if 201 Save power consumption of first server, starting can not be quickly responded according to work requirements after shut down, lead to energy conservation
While starting efficiency it is low, influence first server 201 provide data service.
Step 102: when determining that the first server is in idling conditions, Xiang Suoshu first server sends shutdown life
It enables.
In the present embodiment, cluster management controller 204 monitors each node in cluster in real time, hence, it can be determined that the
The working condition that one server 201 is presently in.
It is that first server 201 is in normal operating conditions for example, first server 201 is in s0 state, at this time
The power consumption of one server 201 is about in its rated disspation or so;First server 201 is in s5 state, at this time first server
201 all building blocks including power supply are all closed, and 201 power consumption of first server is 0;First server 201 is in
Idle IDLE state, the power consumption of first server 201 is greater than 0 and is less than rated disspation at this time.
When being in idling conditions due to first server 201, still there is no small power consumption, therefore, in the present embodiment
By executing step 102, so that first server 201 closes all building blocks of first server 201 according to shutdown command
It closes.That is, switching to s5 state from idling conditions.After switching to s5 state, compared to idling conditions, it is clear that first server 201
Power consumption realization is greatly lowered.
More than, server power-economizing method as provided by the embodiments of the present application, from cluster management controller to first service
Device sends the first control command, and first server stores the initiation parameter of first server memory according to the first control command
Into nonvolatile random access memory NVRAM.When cluster management controller determines that first server is in idling conditions, to
First server sends shutdown command, switches to off-mode from idling conditions to control first server, is cut by state
It changes, saves the power consumption of first server.
Since NVRAM is still able to maintain data after a loss of power, when first server needs the weight under off-mode
When new starting, first server can directly read the initiation parameter of memory from NVRAM, effectively shorten first service in this way
The initialization time of device memory promotes the starting speed of first server.It can be seen that the power-economizing method is not only energy saving, may be used also
So that server quick start after energy conservation shutdown, promotes the working efficiency of server.
It is understood that above embodiments are only described using first server as example, it is in practical applications, right
Any one server in cluster can carry out energy conservation according to the above method.
In some scenes, if the number of servers that data center is in normal operating conditions is excessive, data center's tool
Have and energy-efficient demand is carried out to the server of part normal operating conditions.Based on this demand, the application is still further provided separately
A kind of server power-economizing method.The specific implementation of this method is described in detail below with reference to embodiment and attached drawing.
Second embodiment
Referring to Fig. 3, which is the flow chart of another server power-economizing method provided by the embodiments of the present application.This method is answered
For cluster management controller.About the application scenarios of the present embodiment method, schematic diagram of a scenario shown in Fig. 2 can refer to.
As shown in figure 3, server power-economizing method provided in this embodiment, comprising:
Step 301: Xiang Suoshu first server sends the first control command;First control command is used to indicate first
Server stores the initiation parameter of the memory of the first server into nonvolatile random access memory NVRAM.
The implementation of this step is identical as the implementation of step 101 in previous embodiment, the correlation about step 301
Description can be found in previous embodiment, and details are not described herein again.
Step 302: obtaining the corresponding data of each server in cluster.
It is understood that first server belongs to a node of cluster, therefore each server includes in this step
First server.Since cluster management controller can be transmitted with each server communication and data, can obtain each
The corresponding data of server.
Herein, the corresponding data of each server, can include but is not limited to: the ongoing task quantity of server,
Task names, task definition, EMS memory occupation amount of each task etc..
Step 303: analyzing the corresponding data of each server in the cluster, the task of the first server is shifted
To other servers in each server.
In the present embodiment, cluster management controller has data analysis function.By to the corresponding data of each server
It is analyzed, is capable of forming the strategy of task transfer.For example, first server, second server and third server into
Capable task quantity is respectively 5,35 and 20.Cluster management controller is analyzed and is calculated by data, is determined first
The ongoing task of server is transferred on second server and/or third server and carries out, and first server is switched
For s5 state, the power consumption of data center can be effectively saved.At this point, a task transition strategy can be formed, i.e., by
The ongoing task of one server is transferred to second server and/or third server.
It after the strategy execution, is shifted by task, tentatively first server can be made to be in idle condition.
It is understood that the above is only task transfer examples to be switched to IDLE state from s0 state in practical applications
Number of servers can be one, or it is multiple.In the present embodiment, only it is described by example of first server,
The specific processor for carrying out state switching for application the present embodiment method is not limited.
Step 304: when determining that the first server is in idling conditions, Xiang Suoshu first server sends shutdown life
It enables.
In the present embodiment, the implementation of step 304 is identical as the implementation of step 102 in previous embodiment, about
The associated description of step 304 can refer to previous embodiment, and details are not described herein again.
Step 305: obtaining the mission requirements in cluster.
By executing above step 301 to 304, realize first server from s0 state to the switching of s5 state.This process
Save the power consumption of first server.
In practical applications, data center is also possible to face the case where task amount increases sharply, at this point, data center is in normal
The server of working condition s0 can no longer meet current task processing requirements.Therefore, it is necessary to will be in the one of s5 state
Or multiple servers are switched to s0 state, quick start and can normally handle task.
In this step, the mission requirements in cluster are obtained, the purpose is to determinations to need the server under several s5 states
Switch to s0 state.For ease of understanding, still using first server as example, its stateful switchover process is carried out in conjunction with step 306
Description.
Step 306: according to the mission requirements, Xiang Suoshu first server sends start command, so as to first clothes
Device be engaged in response to the start command, starting is completed after reading the initiation parameter in the NVRAM.
In this step, according to mission requirements, cluster management controller determines the first server needed to s5 state is in
Carry out state switching.At this point, sending start command from cluster management controller to first server.When first server receives
After the start command, that is, the order is responded, the first server memory once stored when read step 301 executes from NVRAM
Initiation parameter.First server easily carries out the physical layer of memory and the parameter of link layer is matched according to initiation parameter
It sets.
The initialization that memory is directly carried out by reading initiation parameter, accelerates the launching process of first server, takes
Be engaged in device can quick response data center task processing requirements.As it can be seen that first server switches to s0 state from s5 state
Speed is exceedingly fast.
It is above server power-economizing method provided by the embodiments of the present application, this method completes first server from s0
To the switching of s5 state, and from s5 to s0 state switching.In the handoff procedure of the state from s0 to s5, server section is realized
Can, reduce total power consumption of data center;And from s5 to s0 in the handoff procedure of state, it is stored in advance in by reading
The internal memory initialization parameter of NVRAM, effectively improves speed of response of server.The power save mode is able to ascend the sexual valence of server
Than and competitiveness.
Based on the server power-economizing method that previous embodiment provides, correspondingly, the application also provides a kind of cluster management control
Device processed.The specific implementation of the cluster management controller is described below with reference to embodiment and attached drawing.
3rd embodiment
Referring to fig. 4, which is the structural schematic diagram of cluster management controller provided by the embodiments of the present application.
As shown in figure 4, cluster management controller provided in this embodiment, comprising:
First sending module 401, for sending the first control command to the first server;First control command
It is used to indicate first server and stores the initiation parameter of the memory of the first server to based non-volatile random access and deposit
In reservoir NVRAM;The first server and the NVRAM are connected with each other;
Second sending module 402, when for determining that the first server is in idling conditions, Xiang Suoshu first server
Send shutdown command.
Since NVRAM is still able to maintain data after a loss of power, cluster management controller can control first service
Device directly reads the initiation parameter of memory from NVRAM, effectively shortens in this way when needing to restart under off-mode
The initialization time of first server memory promotes the starting speed of first server.It can be seen that the cluster management controller
It is not only able to control first server energy conservation, can also control makes server quick start after energy conservation shutdown, mentions to help
Rise the working efficiency of server.
In some scenes, if the number of servers that data center is in normal operating conditions is excessive, data center's tool
Have and energy-efficient demand is carried out to the server of part normal operating conditions.Based on this demand, optionally, cluster management controller is also
Include:
First obtains module, for obtaining the corresponding data of each server in cluster;Each server includes institute
State first server;
Data analysis module, for analyzing the corresponding data of each server in the cluster;
Task shift module, other clothes for being transferred to the task of the first server in each server
Business device.
Optionally, cluster management controller further include:
Second obtains module, for obtaining the mission requirements in cluster;
Third sending module, for according to the mission requirements, Xiang Suoshu first server to send start command, with toilet
First server is stated in response to the start command, starting is completed after reading the initiation parameter in the NVRAM.
As it can be seen that cluster management controller can control the switching so that first server completion state from s0 to s5, and
The switching of state from s5 to s0.In the handoff procedure of the state from s0 to s5, control server realizes energy conservation, reduces data center
Total power consumption;And from s5 to s0 in the handoff procedure of state, NVRAM is stored in advance in by control first server reading
Internal memory initialization parameter, effectively improve speed of response of server.The cluster management controller helps to make with hop server
With and promote the cost performance and competitiveness of server.
Based on aforementioned first embodiment, correspondingly, the application also provides another service by executing subject of server
Device power-economizing method.Below in an example, executing subject server can be the first server in previous embodiment.Below
This method is described with attached drawing in conjunction with the embodiments.
Fourth embodiment
Referring to Fig. 5, which is another server power-economizing method provided by the embodiments of the present application.
As shown in figure 5, the present embodiment method includes:
Step 501: receiving the first control command.
First control command is used to indicate server and stores the initiation parameter of the memory of the server to non-volatile
In property random access storage device NVRAM.The server and the NVRAM are connected with each other.
Step 502: being stored the initiation parameter of the memory of the server to non-easy according to first control command
In the property lost random access storage device NVRAM.
Step 503: shutdown command is received when the server is in idling conditions, it will be described according to the shutdown command
All building blocks of server are closed.
Server power-economizing method provided by the embodiments of the present application sends the first control from cluster management controller to server
The initiation parameter of server memory is stored to based non-volatile random access according to the first control command and is stored by order, server
In device NVRAM.When cluster management controller determines that server is in idling conditions, shutdown command is sent to server, to control
Control server switches to off-mode from idling conditions, is switched by state, saves the power consumption of server.
Since NVRAM is still able to maintain data after a loss of power, when server needs to open again under off-mode
When dynamic, server can directly read the initiation parameter of memory from NVRAM, effectively shorten the initial of server memory in this way
Change the time, promotes the starting speed of server.It can be seen that the power-economizing method is not only energy saving, server can also be made in energy conservation
Quick start after shutdown promotes the working efficiency of server.
In some scenes, if the number of servers that data center is in normal operating conditions is excessive, data center's tool
Have and energy-efficient demand is carried out to the server of part normal operating conditions.Based on this demand, the application still further provide and
A kind of server power-economizing method, is still described by executing subject of server.It should be noted that the present embodiment and second is in fact
Example reciprocal correspondence is applied, only executing subject is distinct, and the associated description about the present embodiment can refer to second embodiment.It ties below
Embodiment and attached drawing are closed the specific implementation of this method is described.
5th embodiment
Referring to Fig. 6, which is the flow chart of another server power-economizing method provided by the embodiments of the present application.This method is answered
For server.About the application scenarios of the present embodiment method, schematic diagram of a scenario shown in Fig. 2 can refer to.
As shown in fig. 6, server power-economizing method provided in this embodiment, comprising:
Step 601: the corresponding data of the server are sent to cluster management controller, so as to cluster management control
Device is analyzed according to the corresponding data of the server, and the task of the server is transferred to other services in cluster
Device.
Step 602: receiving the first control command.
First control command is used to indicate server and stores the initiation parameter of the memory of the server to non-volatile
In property random access storage device NVRAM.The server and the NVRAM are connected with each other.
Step 603: being stored the initiation parameter of the memory of the server to non-easy according to first control command
In the property lost random access storage device NVRAM.
Step 604: shutdown command is received when the server is in idling conditions, it will be described according to the shutdown command
All building blocks of server are closed.
Step 605: receiving start command.
In the present embodiment, start command is used to indicate the server starting.
Step 606: in response to the start command, the initiation parameter, the server are read from the NVRAM
Starting.
It is above server power-economizing method provided by the embodiments of the present application, this method completes first server from s0
To the switching of s5 state, and from s5 to s0 state switching.In the handoff procedure of the state from s0 to s5, server section is realized
Can, reduce total power consumption of data center;And from s5 to s0 in the handoff procedure of state, it is stored in advance in by reading
The internal memory initialization parameter of NVRAM, effectively improves speed of response of server.The power save mode is able to ascend the sexual valence of server
Than and competitiveness.
Based on aforementioned fourth embodiment and the 5th embodiment, correspondingly, the application also provides a kind of energy-saving server.Below
The specific implementation of the server is described with attached drawing in conjunction with the embodiments.
Sixth embodiment
Referring to Fig. 7, which is a kind of structural schematic diagram of energy-saving server provided by the embodiments of the present application.It needs to illustrate
It is that energy-saving server is specifically as follows the aforementioned first first server described to 3rd embodiment in the present embodiment.
As shown in fig. 7, the energy-saving server includes:
First receiving unit 701, for receiving the first control command;
Data storage cell 702, for being joined the initialization of the memory of the server according to first control command
Number is stored into nonvolatile random access memory NVRAM;The server and the NVRAM are connected with each other;
Second receiving unit 703, for receiving shutdown command when the server is in idling conditions;
Shut down unit 704, for being closed all building blocks of the server according to the shutdown command.
In some scenes, if the number of servers that data center is in normal operating conditions is excessive, data center's tool
Have and energy-efficient demand is carried out to the server of part normal operating conditions.Based on this demand, optionally, server further include:
First transmission unit, for sending the corresponding data of the server to cluster management controller, so as to the collection
Group's Management Controller is analyzed according to the corresponding data of the server, and the task of the server is transferred in cluster
Other servers.
Optionally, server further include:
Third receiving unit, for receiving start command;The start command is used to indicate the server starting;
Start unit, it is described for reading the initiation parameter from the NVRAM in response to the start command
Server starting.
Server can complete the switching of the state from s0 to s5, and from s5 to s0 state switching.In the shape from s0 to s5
In the handoff procedure of state, server realizes energy conservation, reduces total power consumption of data center;And from s5 to s0 state switching
Cheng Zhong, server are stored in advance in the internal memory initialization parameter of NVRAM by reading, effectively improve speed of response of server.
As it can be seen that server provided in this embodiment cost performance with higher and competitiveness.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment it
Between same and similar part may refer to each other, each embodiment focuses on the differences from other embodiments.
For equipment and system embodiment, since it is substantially similar to the method embodiment, so describe fairly simple,
The relevent part can refer to the partial explaination of embodiments of method.Equipment and system embodiment described above is only schematic
, wherein unit may or may not be physically separated as illustrated by the separation member, as unit prompt
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.Some or all of the modules therein can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
Those of ordinary skill in the art can understand and implement without creative efforts.
The above, only a kind of specific embodiment of the application, but the protection scope of the application is not limited thereto,
Within the technical scope of the present application, any changes or substitutions that can be easily thought of by anyone skilled in the art,
Should all it cover within the scope of protection of this application.Therefore, the protection scope of the application should be with scope of protection of the claims
Subject to.
Claims (12)
1. a kind of server power-economizing method, which is characterized in that be applied to cluster management controller, which comprises
The first control command is sent to the first server;First control command is used to indicate first server will be described
The initiation parameter of the memory of first server is stored into nonvolatile random access memory NVRAM;The first service
Device and the NVRAM are connected with each other;
When determining that the first server is in idling conditions, Xiang Suoshu first server sends shutdown command.
2. the method according to claim 1, wherein the determination server be in idling conditions it
Before, the method also includes:
Obtain the corresponding data of each server in cluster;Each server includes the first server;
The corresponding data of each server in the cluster are analyzed, the task of the first server is transferred to each clothes
Other servers being engaged in device.
3. method according to claim 1 or 2, which is characterized in that send shutdown life to the first server described
After order, the method also includes:
Obtain the mission requirements in cluster;
According to the mission requirements, Xiang Suoshu first server sends start command, so that the first server is in response to institute
Start command is stated, starting is completed after reading the initiation parameter in the NVRAM.
4. a kind of cluster management controller characterized by comprising
First sending module, for sending the first control command to the first server;First control command is for referring to
Show that first server stores the initiation parameter of the memory of the first server to nonvolatile random access memory
In NVRAM;The first server and the NVRAM are connected with each other;
Second sending module, when for determining that the first server is in idling conditions, Xiang Suoshu first server, which is sent, is closed
Machine order.
5. cluster management controller according to claim 4, which is characterized in that further include:
First obtains module, for obtaining the corresponding data of each server in cluster;Each server includes described
One server;
Data analysis module, for analyzing the corresponding data of each server in the cluster;
Task shift module, other services for being transferred to the task of the first server in each server
Device.
6. cluster management controller according to claim 4 or 5, which is characterized in that further include:
Second obtains module, for obtaining the mission requirements in cluster;
Third sending module, for according to the mission requirements, Xiang Suoshu first server to send start command, so as to described the
One server completes starting in response to the start command after reading the initiation parameter in the NVRAM.
7. a kind of server power-economizing method, which is characterized in that be applied to server, which comprises
Receive the first control command;
The initiation parameter of the memory of the server is stored to based non-volatile random access according to first control command
In memory NVRAM;The server and the NVRAM are connected with each other;
Shutdown command is received when the server is in idling conditions, according to the shutdown command owning the server
Building block is closed.
8. the method according to the description of claim 7 is characterized in that before the server is in idling conditions, the side
Method further include:
The corresponding data of the server are sent to cluster management controller, so that the cluster management controller is according to the clothes
The corresponding data of business device are analyzed, and the task of the server is transferred to other servers in cluster.
9. method according to claim 7 or 8, which is characterized in that it is described according to the shutdown command by the service
After all building blocks of device are closed, the method also includes:
Receive start command;The start command is used to indicate the server starting;
In response to the start command, the initiation parameter, the server starting are read from the NVRAM.
10. a kind of energy-saving server characterized by comprising
First receiving unit, for receiving the first control command;
Data storage cell, for according to first control command by the initiation parameter of the memory of the server store to
In nonvolatile random access memory NVRAM;The server and the NVRAM are connected with each other;
Second receiving unit, for receiving shutdown command when the server is in idling conditions;
Shut down unit, for being closed all building blocks of the server according to the shutdown command.
11. server according to claim 10, which is characterized in that further include:
First transmission unit, for sending the corresponding data of the server to cluster management controller, so as to the cluster pipe
Reason controller is analyzed according to the corresponding data of the server, and the task of the server is transferred to its in cluster
His server.
12. server described in 0 or 11 according to claim 1, which is characterized in that further include:
Third receiving unit, for receiving start command;The start command is used to indicate the server starting;
Start unit, for reading the initiation parameter, the service from the NVRAM in response to the start command
Device starting.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910213121.7A CN109917902A (en) | 2019-03-20 | 2019-03-20 | A kind of server power-economizing method, cluster management controller and energy-saving server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910213121.7A CN109917902A (en) | 2019-03-20 | 2019-03-20 | A kind of server power-economizing method, cluster management controller and energy-saving server |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109917902A true CN109917902A (en) | 2019-06-21 |
Family
ID=66965910
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910213121.7A Pending CN109917902A (en) | 2019-03-20 | 2019-03-20 | A kind of server power-economizing method, cluster management controller and energy-saving server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109917902A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110703900A (en) * | 2019-09-23 | 2020-01-17 | 苏州浪潮智能科技有限公司 | Power control method and device for cloud computing environment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101937344A (en) * | 2009-06-30 | 2011-01-05 | 联想(北京)有限公司 | Computer and method for quickly starting same |
CN104270430A (en) * | 2014-09-19 | 2015-01-07 | 国云科技股份有限公司 | Server remote dispatching method suitable for cloud computing |
CN105117246A (en) * | 2015-08-04 | 2015-12-02 | 晶晨半导体(上海)有限公司 | Method for rapidly booting electronic equipment |
CN105975047A (en) * | 2016-04-29 | 2016-09-28 | 北京百度网讯科技有限公司 | Method and system for adjusting power consumption of cloud data centre |
CN106126206A (en) * | 2016-06-15 | 2016-11-16 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
CN109324875A (en) * | 2018-09-27 | 2019-02-12 | 杭州电子科技大学 | A kind of data center server power managed and optimization method based on intensified learning |
-
2019
- 2019-03-20 CN CN201910213121.7A patent/CN109917902A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101937344A (en) * | 2009-06-30 | 2011-01-05 | 联想(北京)有限公司 | Computer and method for quickly starting same |
CN104270430A (en) * | 2014-09-19 | 2015-01-07 | 国云科技股份有限公司 | Server remote dispatching method suitable for cloud computing |
CN105117246A (en) * | 2015-08-04 | 2015-12-02 | 晶晨半导体(上海)有限公司 | Method for rapidly booting electronic equipment |
CN105975047A (en) * | 2016-04-29 | 2016-09-28 | 北京百度网讯科技有限公司 | Method and system for adjusting power consumption of cloud data centre |
CN106126206A (en) * | 2016-06-15 | 2016-11-16 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
CN109324875A (en) * | 2018-09-27 | 2019-02-12 | 杭州电子科技大学 | A kind of data center server power managed and optimization method based on intensified learning |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110703900A (en) * | 2019-09-23 | 2020-01-17 | 苏州浪潮智能科技有限公司 | Power control method and device for cloud computing environment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110413392B (en) | Method for formulating single task migration strategy in mobile edge computing scene | |
Bolla et al. | Enabling backbone networks to sleep | |
Jin et al. | Joint host-network optimization for energy-efficient data center networking | |
US9632839B2 (en) | Dynamic virtual machine consolidation | |
Xia et al. | Phone2Cloud: Exploiting computation offloading for energy saving on smartphones in mobile cloud computing | |
Bolla et al. | Energy efficiency in the future internet: A survey of existing approaches and trends in energy-aware fixed network infrastructures | |
Nguyen et al. | Powering a data center network via renewable energy: A green testbed | |
Rahman et al. | Energy saving in mobile cloud computing | |
CN102317914B (en) | Methods, system and devices for managing virtual resources | |
CN103077082A (en) | Method and system for distributing data center load and saving energy during virtual machine migration | |
CN101986679A (en) | Mobile terminal, electricity-saving method and electricity-saving device | |
RU2741757C1 (en) | Method of resuming multiple connection in sleep state and device for realizing said method | |
Xiong et al. | Lightpath management in SDN-based elastic optical networks with power consumption considerations | |
Boru et al. | Models for efficient data replication in cloud computing datacenters | |
CN108664116A (en) | Adaptive electricity saving method, device and the cpu controller of network function virtualization | |
US20140215041A1 (en) | Workload migration determination at multiple compute hierarchy levels | |
US20170116094A1 (en) | Fault handling methods in a home service system, and associated household appliances and servers | |
CN111404818B (en) | Routing protocol optimization method for general multi-core network processor | |
CN109917902A (en) | A kind of server power-economizing method, cluster management controller and energy-saving server | |
Zhu et al. | Energy efficient dynamic virtual optical network embedding in sliceable-transponder-equipped EONs | |
Kaur et al. | Improved distributed fault tolerant clustering algorithm for fault tolerance in WSN | |
CN113630481B (en) | Automatic in-band control plane construction method and system in SDN | |
CN115374949A (en) | Distributed quantum computing system and resource management method | |
CN102521046B (en) | Server and working method thereof | |
CN114077213B (en) | Energy-saving control method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190621 |